Author Login Editor-in-Chief Peer Review Editor Work Office Work

05 May 2012, Volume 38 Issue 9
    

  • Select all
    |
    Networks and Communications
  • LI Xi- , DAN Chang-Min, LI Chang, CHEN Feng-Dui, TIAN Li-Jiao
    Computer Engineering. 2012, 38(9): 1-3. https://doi.org/10.3969/j.issn.1000-3428.2012.09.001
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Spectral unmixing of remote sensing images is a hotspot in remote sensing field, and Multilayer Perception(MLP) neural network is a common nonlinear spectral unmixing algorithm. However, currently there is no effective way to deal with the negative abundances derived by the network. To solve this problem, a MLP neural network with variable architecture is proposed. By discarding endmembers with negative abundances, the MLP architecture is modified to unmix the rest endmembers, so a remote sensing image is finally unmixed. An experiment using a simulated TM image shows that the average errors of the proposed method, conventional MLP method and linear spectral unmixing model are 0.077 7, 0.081 9 and 0.094 3 respectively, thus the proposed method outperforms the other two. Therefore, the proposed method can overcome the negative abundance problem effectively.
  • WANG En-Cheng, SU Teng-Fang, YUAN Kai-Guo, WU Chun-Hua, WANG Yu-Qiang
    Computer Engineering. 2012, 38(9): 4-7. https://doi.org/10.3969/j.issn.1000-3428.2012.09.002
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to improve the retrieval performance of using melody contour and rhythm in Query By Humming(QBH), this paper proposes a new algorithm using pitch and energy for note segmentation. Autocorrelation pitch extraction algorithm is improved. Post processing is used to smooth the pitch frequency curves, and at the same time, energy segmentation information is maintained in the process of note segmentation. Semitone curve is used for segmentation. This algorithm improves the accuracy of the note segmentation. Experimental results indicate that the accuracy rate of the algorithm is 88.75% in quiet laboratory environment which is of great significance in QBH.
  • ZHOU Gong-Fang, LI Gong-Yan, LIU Ying, WANG Xiao-Dong
    Computer Engineering. 2012, 38(9): 8-11. https://doi.org/10.3969/j.issn.1000-3428.2012.09.003
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to better determine the optimal clustering number for multi-dimensional data, this paper proposes an new algorithm——MHC, which is based on the principle of the traditional algorithm to determine the clustering number for the dataset. This algorithm adopts bottom-up method to generate dataset partition of different levels. In every division, the algorithm automatically generates the partition of clustering quality, and chooses the optimal clustering number by the clustering quality. Additionally, it still presents a new clustering validity index——Between-In- Proportion(BIP), which is used to measure the different division of clustering quality, and mainly depends on the geometrical structure of datasets. Theoretical analysis and experimental results verify the effectiveness and good performance of the new validity index and the MHC algorithm.
  • TAO Yao, WANG Xin-Wei
    Computer Engineering. 2012, 38(9): 12-14. https://doi.org/10.3969/j.issn.1000-3428.2012.09.004
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In this paper, a new protocol which combines cross-domain authentication and key agreement together is proposed to solve the repeat computing problem in both procedures. Utilizing public key cryptography, the proposed protocol combines both cross-domain authentication protocol and key agreement protocol, and implements both functions in a single protocol with less computational resources, storage resources and network bandwidth. Theoretical analysis and performance comparison results show that the proposed protocol can provide confidentiality, authentication, integrity, nonrepudiation and other security guarantees. Besides, it has a high level of efficiency.
  • YANG Bei, XIAO Zi-Bi, LI Shou-Gui, SONG Chun-Guang
    Computer Engineering. 2012, 38(9): 15-18. https://doi.org/10.3969/j.issn.1000-3428.2012.09.005
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper analyzes a certificateless signature scheme based on discrete logarithm problem, which is proposed by Zhang Yanyan et al and points out that their scheme is universally forgeable, and an adversary can forge any signer’s valid signature on any message by replacing public key. It also presents an improved scheme which is proved to be existentially unforgeable in a super security model. In addition, the improved scheme has the advantage of high computation efficiency, as there is no heavily cost bilinear pairing operation in the improved scheme.
  • YANG Yang, CHENG Nan-Nan, ZHANG Meng-Meng
    Computer Engineering. 2012, 38(9): 19-20,35. https://doi.org/10.3969/j.issn.1000-3428.2012.09.006
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    According to the requirement of hand posture control system in the field of human machine interaction, Average Neighborhood Margin Maximization(ANMM) algorithm is applied to static hand posture and human body posture recognition. It normalizes the binary image contour into a fixed resolution, constitutes a multi-dimensional vector, uses the ANMM algorithm training the homogeneity and heterogeneity neighboring vectors, and then projection direction matrix is calculated. After reducing the dimension of the sample, calculating the distance of the adjacent homogeneity and heterogeneity vectors, the samples are classified. Results show that this algorithm has a recognition rate of 90% for the static hand posture recognition.
  • LIU Xiao-Shuang, SHUAI Jian-Mei, TUN Qiang-Xiang
    Computer Engineering. 2012, 38(9): 21-23,27. https://doi.org/10.3969/j.issn.1000-3428.2012.09.007
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper presents a novel memory manager for object-oriented programs. It reduces memory fragmentation by using a hybrid data structure to manage small objects, medium objects and big objects. It reduces the time complexity of memory allocation and de-allocation through segregated fit and bitmap fit algorithms. Coalescence of free chunks is fast due to the use of boundary tag technology and response speed is accelerated by taking advantage of the cache technology and program locality. Experimental results show that the memory manager has quicker response speed and higher memory usage.
  • HU Di, LV Guo-Nian, LIN Ba-Gong, WEN Yong-Ning
    Computer Engineering. 2012, 38(9): 24-27. https://doi.org/10.3969/j.issn.1000-3428.2012.09.008
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Based on analysis of the content, structure and features of Zhujuan resumes, a text entry data model and a spatial-temporal data model of Zhujuan resume are designed for data organization and management, with the characteristics of clear classification, fine-grained information, locatable time and locatable space. The data collection of Zhujuan resume can be divided into data collection of text entries and conversion from text entries into spatial-temporal data model. The family tree GIS platform is employed for splitting and parsing text entries. Experimental result shows that the method can effectively realize the digitization and spatialization.
  • TAO Yuan, DIAO Rong-Cai
    Computer Engineering. 2012, 38(9): 28-31. https://doi.org/10.3969/j.issn.1000-3428.2012.09.009
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    On the basis of Open64 compiler framework, this paper proposes an algorithm which can implement the continuous reference analysis in nest loops based on profile information and corresponding vectorization optimization method. By using the feed-back compiling optimization techniques, the algorithm can obtain runtime profile information about the continuity of the program and implement loop vectorization by structure peeling and data reorganization. Experimental results show that the algorithm can provide more accurate vectorization information for the irregular code and improve the vectorization extent of code.
  • TAN Zheng, XU Jian-Min, LIN Bao-Li, ZHUO Qing-Hua
    Computer Engineering. 2012, 38(9): 32-35. https://doi.org/10.3969/j.issn.1000-3428.2012.09.010
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes a long-distance driving test monitoring system of 3-lay architecture. This system collects the drivers’ driving information by driving recorder. It establishes a model to judge the drivers’ test performance through matching route, speed, driving time etc, builds the oracle database to save relevant data for enquiries and statistics. Test result shows that this system efficiently assess driver’s actual driving ability, detectes the illegal activities of long distance driving test.
  • LI Jing-Bo, LIU Jin-Gang, GENG Yun-Bing
    Computer Engineering. 2012, 38(9): 36-39. https://doi.org/10.3969/j.issn.1000-3428.2012.09.011
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problem that existing business service selection method has low efficiency, this paper proposes an improved service selection method, which establishes the inverted index between the features of business services and physical services. During the stage of business service modeling, it can synchronize the modeling operations and physical service binding. At the stage of business service runtime, it can synchronize the feature selection and physical service selection. Experimental result shows it reduces the operating time and improves the availability of business service.
  • WANG Qiang-Dan, JIANG Chu-Juan, ZHANG Yan-Mei
    Computer Engineering. 2012, 38(9): 40-42. https://doi.org/10.3969/j.issn.1000-3428.2012.09.012
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes an improved method to generate the feasible basis paths, taking into account the influence of branch correlation and exception-handling constructs on the feasibility of paths. It analyzes the causes of generation for infeasible paths and gives its detecting method, constructs exceptional control flow graph which computes the branch correlations to determine the feasibility of paths, builds the feasible basis path set. Case analysis results show that the method can effectively avoid infeasible path in basis path set, and it meets the requirements of the basis path test.
  • JI Xiao-Dong, HUA Qiang-Yi, TUN Hao, LI E-Meng, GUO Cui, CHANG Xiao-Jun
    Computer Engineering. 2012, 38(9): 43-45. https://doi.org/10.3969/j.issn.1000-3428.2012.09.013
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The current model-based user interface design is not a very good solution for the isolation between interface and semantic. This paper proposes a design of model-based user interface transformer, which divides the design process into five hierarchies that are task and domain, interface design, abstract user interface, concrete user interface and final user interface. It makes use of the associated user interface description language to implement abstract description, and exports the abstract model of user interface. After layer-by-layer model mapping, ultimately parses and displays it by the client. Application results show that the method has good usability, and it can fulfill the purpose of cross-platform.
  • WANG Qi, LI Xia
    Computer Engineering. 2012, 38(9): 46-48. https://doi.org/10.3969/j.issn.1000-3428.2012.09.014
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes a classification rule reduction method by analyzing the correlation of attributes in classification rules. It obtains the association rule set by analyzing the correlation among the attributes of training set, describes the importance degree in the classification rule by using the degree of confidence α of the association rule. The later part of the association rule, whose α is larger than threshold value ?, is deleted in the classification rule. Experimental results validate that this method has higher classification effectiveness by using UCI and SDSS data as the decision system.
  • RU Bei, XIAO Yun-Feng, ZHANG Dun-Feng
    Computer Engineering. 2012, 38(9): 49-50,54. https://doi.org/10.3969/j.issn.1000-3428.2012.09.015
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper combines with Aglets platform and J2EE servlet technology, proposes a Java Platform Mobile Edition(J2ME) Web service integration solutions of three layers based on Agent. In the terminal layer, lightweight proxy access methods are used to reduce the carrying needs of resource-constrained mobile device. In web access layer, the Web service standard access method is used to ensure heterogeneous mobile platform unified access. In mobile Agent layer, multiply Agents working together ensure system efficiency and flexibility. On the basis of it, a mobile stock exchange system is developed. Application result shows that this scheme can improve the efficiency and robustness of J2ME equipment discovery, access Web services under wireless environment.
  • GE Jun, ZHOU Lian-Yang
    Computer Engineering. 2012, 38(9): 51-54. https://doi.org/10.3969/j.issn.1000-3428.2012.09.016
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at many inadequacies of the existing sandstorm forecasting systems, this paper studies the connectivity issue of Wireless Sensor Network(WSN) for sandstorm monitoring. Taking into account such channel has the characteristics of diversity/variability under the impact of sandstorms, the paper gives multiple types of channels analysis method based on percolation theory. It proves that jointly using multiple types of channels improves network connectivity, and proposes a feasible WSN access switching method. Study results show that the WSN architecture based on the proposed access switching method can construct an efficient sandstorm monitoring system, and provide an effective control to sandstorm.
  • XU Guang-Xian, FU Xiao
    Computer Engineering. 2012, 38(9): 55-57. https://doi.org/10.3969/j.issn.1000-3428.2012.09.017
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the high complexity of coding of secure network, this paper proposes a secure network coding algorithm based on sparse matrix. The sparse matrix has a fast computing speed and occupies smaller memory space, so the algorithm transforms the information source news with sparse matrix. The algorithm can achieve the information-theoretic security condition at a high probability. Simulation results confirm that the algorithm can increase the encoding and decoding rate, reduce the computational complexity and the memory space.
  • WU Chun-Hua, LIU Yi
    Computer Engineering. 2012, 38(9): 58-61. https://doi.org/10.3969/j.issn.1000-3428.2012.09.018
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    On the basis of Wireless Sensor Network(WSN) Ad Hoc On-demand Distance Vector(AODV) routing protocol, this paper proposes a WSN on-demand multipath routing protocol based on energy-aware, named AODV-EALB. It combines with energy link quality and forwarding priority factor. Through the consideration of the surplus energy and signal strength of node acceptging packets, it balances the network node energy consumption. Simulation experimental results show that AODV-EALB protocol can reduce the packet loss rate and extend the network lifetime.
  • GAO Ti-Ti, WANG Rui, MO Wang-Gen, TU Shuai
    Computer Engineering. 2012, 38(9): 62-65,68. https://doi.org/10.3969/j.issn.1000-3428.2012.09.019
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper studies the fuzzy information exposure paths in Wireless Sensor Network(WSN), and presents an algorithm to dispose the fuzzy information in both of optimal and worst way. It builds the fuzzy coverage model based on the sensor coverage radius, the fusion parameters and coverage threshold. The influence from the fusion factor and parameter of coverage intensity on path information exposure is analyzed. Simulation results show that compared with the probability coverage model, the fuzzy coverage model can get less information exposure intensity on the worse path, while more intensity on the optimal path.
  • DENG Hui-Beng, TUN Yu-Cheng, QI Mei-Juan, TUN Jin-Cheng
    Computer Engineering. 2012, 38(9): 66-68. https://doi.org/10.3969/j.issn.1000-3428.2012.09.020
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to improve the diversity performance of Double Space-time Transmit Diversity(DSTTD) system in Independent and Identically Distributed(IID) fading channels, a DSTTD system combined with Signal Space Diversity(SSD) is proposed. The system can acquire diversity gain by rotation mapping and I-Q branch interleaver. Meanwhile, based on the combined detection method of transmit model system by matrix transformation, the system performance is largely improved by using vector Gaussian approximation and scalar Gaussian approximation for iterative decoding. Simulation results show that the system proposed has 2 dB gain than the ordinary DSTTD system in IID fading channel.
  • CHEN Jiang, YANG Xiao-Feng, YANG Dong
    Computer Engineering. 2012, 38(9): 69-71,75. https://doi.org/10.3969/j.issn.1000-3428.2012.09.021
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper analyzes the multipath delay, Doppler effect and other transmission characteristics of the low-altitude wireless channel, uses Wide Sense Stationary Uncorrelated Scattering(WSSUS) model, according to the four different scenarios of parking, en-routing, arriving and over the top models are respectively set up. Simulation experiment of this model is carried out by using the Monte Carlo(MC) method, and results show that the channel model to build a true reflection of the characteristics of low-altitude wireless channel.
  • JIANG Hong-Lu, YU Ji-Guo, WANG Guang-Hui
    Computer Engineering. 2012, 38(9): 72-75. https://doi.org/10.3969/j.issn.1000-3428.2012.09.022
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper studies the construction of k-broadcast and k-multicast topology control problems in wireless Ad Hoc networks. The goal is to achieve the k-fault-tolerance with a minimal overall power assignment. For the two different topology control mechanism, a new framework of approximation algorithm is proposed to maintain the k-fault-tolerant topology. Simulation experiments show the algorithm has a smaller approximation ratio of O(k) and the total energy consumption is smaller than existing results.
  • CA Chen-Guang, JIANG Xiu-Zhu
    Computer Engineering. 2012, 38(9): 76-78,82. https://doi.org/10.3969/j.issn.1000-3428.2012.09.023
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper analyzes the scheduling principles of Ethernet for Plant Automation(EPA) communication schedule management entity in the EPA micro-segment, discusses the channel schedule efficiency of EPA communication schedule algorithm, and gets the range of channel utilization in periodic data transmission stage and non-periodic data transmission stage under certain conditions. It obtains the maximum and minimum values of channel utilization through derivation. A EPA simulation model is designed, experiment shows that the utilization can obtain the extremum value of theoretical value, and verifies the correctness of the theoretical results.
  • LIU Mei, JIANG Qing-Meng, LIU Si-Qi
    Computer Engineering. 2012, 38(9): 79-82. https://doi.org/10.3969/j.issn.1000-3428.2012.09.024
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Original TCP protocol often misjudges the congestion status and reacts slowly in wireless multi-hop networks, and this paper proposes an implementation method of Semi-TCP protocol. It uses a Virtual TCP(VTCP) layer to control the filter and retransmission of the packets, a logical channel designed to deliver cross-layer information, and a rate adjustment mechanism with one-hop ACK packet. Simulation experimental results show that this method can improve congestion control performance without changing original TCP framework.
  • JIN Shun-Fu, LV Qian, WANG Peng, LI Xiao-Liang
    Computer Engineering. 2012, 38(9): 83-85,92. https://doi.org/10.3969/j.issn.1000-3428.2012.09.025
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at Quality of Service(QoS) requirements of different traffics in the current network, a dynamic bandwidth allocation strategy named E-DBA is proposed by taking into account both the traffic priority and system load. Geom/G/1 queuing model with multiple working vacations mechanism is established. By using the method of embedded Markov chain, the steady-state measures of the queueing model are derived. The performance measures in terms of the average response time and system utility of E-DBA are given. With experiments, the influence of the guaranteed bandwidth of class II traffic on the system performance is illustrated.
  • TAO Yan, YU Ji-Guo, GUO Jiang
    Computer Engineering. 2012, 38(9): 86-89,96. https://doi.org/10.3969/j.issn.1000-3428.2012.09.026
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper analyzes the range-free localization Algorithm APIT and API2T, points out the drawback of low accuracy and poor coverage. Wireless Sensor Network(WSN) localization algorithm based on grid scanning, called API3T, is proposed. In-To-Out error and Out-To-In error are reduced by redefining the neighbor nodes and it improves the strategy of the grid-scanning. Simulation experimental results show that the algorithm increases the localization accuracy and coverage.
  • WANG Shuai, XU Jin-Song, LIANG Pei, TIAN Hua-Wen
    Computer Engineering. 2012, 38(9): 90-92. https://doi.org/10.3969/j.issn.1000-3428.2012.09.027
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to solve the problem of locating unknown nodes in three-dimensional space, this paper proposes a three-dimensional weighed centroid localization algorithm based on tetrahedral model. The tetrahedron consists of four anchor nodes. The unknown nodes are located according to the distances between unknown nodes and anchor nodes, which are weighted sum. It uses the data information of Received Signal Strength Indicator(RSSI) to revise weighting factor. Simulation results show that the localization accuracy of the algorithm is high.
  • LIU Yong-Shuai, WANG Xin-Sheng, HU Yu-Lan
    Computer Engineering. 2012, 38(9): 93-96. https://doi.org/10.3969/j.issn.1000-3428.2012.09.028
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In the multi-hop and clustering hierarchical structure, which is often used in Wireless Sensor Network(WSN) of large-scale and high-density, cluster head node is the only way which the communication must be passed between inner-cluster and extra-cluster, it is also the node that congestion is most easily to happen. In order to solve the network congestion problem, this paper presents a Congestion Control algorithm Based on Virtual Cluster head(CCBVC) to release network congestion, including the establishment of Shortest Path Tree(SPT) in inner-cluster and construction of multiplexing path in extra-cluster. Simulation results show that the algorithm not only can ensure network throughput effectively, but also can shorten data transmission delay relatively.
  • ZHANG Yang-Yang, LIU Jin-Gang, WANG Yu-Wei, YANG Bao-Xin
    Computer Engineering. 2012, 38(9): 97-99. https://doi.org/10.3969/j.issn.1000-3428.2012.09.029
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper designs and implements a per-flow handoff protocol among multiple interfaces in Dual Stack Mobile IPv6(DSMIPv6) based on RFC 5555. This protocol proposes a solution for multi-interface mobile terminal to perform business flow redirection flexibly in IPv4 and IPv6 networks, which utilizes the features of Netfilter, XFRM and policy routing of Linux. It provides more efficient handoff means for intelligent handoff algorithm of higher layers. Experimental results show that flow handoff in DSMIPv6 improves the performance of transmission and takes full advantage of the network resource.
  • LIN Lu-Rong, SHANG Bi-Yu
    Computer Engineering. 2012, 38(9): 100-104. https://doi.org/10.3969/j.issn.1000-3428.2012.09.030
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Multi-stage Stratified Link Routing(MSLR) algorithm is proposed on the basis of COSEN. Here all the sensor nodes are sorted according to the distance from base station to themselves and divided into several stages with a fixed number of nodes per layer. Nodes belong to each stage form one cluster and select one node to be cluster-head in the specified range by the parent cluster-head to gather data of its stage and integrate data received from the lower cluster-head, and transmit them to its superior cluster-head until these data reach the base station. Compared with the schemes of PEGASIS and COSEN through NS3 computer simulation platform, result shows that MSLR shortens the transmission delay greatly, and significantly improves the network lifetime.
  • JIN Yong, BAI Guang-Wei
    Computer Engineering. 2012, 38(9): 105-107. https://doi.org/10.3969/j.issn.1000-3428.2012.09.031
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes an adaptive cooperative Automatic Repeat request(ARQ) mechanism based on cross-layer design for Wireless Multimedia Sensor Network(WMSN), in the hope that the QoS guarantee of data transmission can be greatly improved. The hops of data packets are adjusted adaptively according to the characteristics of the energy consumption with communication distance of the cooperative ARQ. The mechanism adaptively adjusts the maximum retransmission time according to the hops of data packets. The GOP fetching frame model based on the fast forward/backward rate is proposed. Analysis result shows that the proposed mechanism can obtain better quality of multimedia data transmission than traditional ARQ, in terms of frame rate, reliability, real-time performance.
  • GAO Hui-Sheng, LI Xiao-Rui, WANG Sha, ZHAN Jing-Yu
    Computer Engineering. 2012, 38(9): 108-110,113. https://doi.org/10.3969/j.issn.1000-3428.2012.09.032
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    With the network induced time delay and data packet losses in Networked Control System(NCS), the performance of control systems degrades even destabilizes. In practice the states of some systems are unmeasured. This paper designs two state observers to estimate the states and when data packet losses, the state observers can compensate it. Exponential stability theory is used to analyze the stability of closed loop system and the method calculating the control parameter is given. A motor is used to simulate. Simulation result shows that the validity of the proposed method which guarantees the stability of network control system with network induced time-delay and data packet losses.
  • ZHOU Ting, MAO Qian
    Computer Engineering. 2012, 38(9): 111-113. https://doi.org/10.3969/j.issn.1000-3428.2012.09.033
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes an image steganographic algorithm based on direction coding in Discrete Cosine Transform(DCT) domain. The secret image is transformed by 8 × 8 block DCT and quantized by using JPEG standard quantification table, then the DCT coefficients are coding by the trip coding. And the cover image is transformed by 8 × 8 block DCT and quantized by using the improved quantification table, which raises the embedding capacity indirectly. The secret image data are embedded in the low frequency of the cover image’s DCT coefficients by using the steganorgraphy algorithm based on direction coding. Experimental results prove that the steganorgraphic method has a great embedding capacity, makes the image containing secret image has a good visual effect, and it can resist to analysis.
  • DUAN Xue-Feng, GUAN Jian, DING Yong, LIU Yun-Bei
    Computer Engineering. 2012, 38(9): 114-116,120. https://doi.org/10.3969/j.issn.1000-3428.2012.09.034
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    There are some problems in classical image scrambling algorithms such as needing many iterations, insufficient strength of encryption, little key space and weak resistance. Based on multi-group chaotic sequences and matrix vectorizating operator, this paper proposes a new color digital image scrambling algorithm to solve the problems. By vectorizing and scrambling the RGB matrix of color digital image, the information of original image is hidden. Theory analysis and experiment results show that the new algorithm has large key space and sensitive to the key. It can change the grey-scale feature of image, and scrambling effect is good.
  • FU Sai-Yuan, LIU Gong-Shen, LI Jian-Hua
    Computer Engineering. 2012, 38(9): 117-120. https://doi.org/10.3969/j.issn.1000-3428.2012.09.035
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Unified Extensible Firmware Interface(UEFI) faces a grim challenges of malicious code attack. The traditional computer security software can not provide security for the firmware and operating system boot process. In order to solve the problem, this paper designs a malicious code defense system based on UEFI firmware. By using the multi-pattern matching algorithm, a signature detecting engine under UEFI environment is implemented, which provides functionally of malicious code detect, boot option analysis and firmware and operating system kernel backup. Experimental results prove that the system can effectively resist malicious code with small code size and low costs to meet the firmware’s need of flash size and fast boot.
  • DIAO Ti-Hua, HUANG Chu-Mei
    Computer Engineering. 2012, 38(9): 121-124. https://doi.org/10.3969/j.issn.1000-3428.2012.09.036
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Due to the limitations of power, computation capability and storage resources, as well as the more complex security threats the current network faces, a pairings-based key negotiation and authentication protocol for class LEACH routing in Wireless Sensor Network(WSN) is proposed, which achieves bidirectional authentication between nodes by introducing the current round of routing to defend familiar attacks such as selective forwarding, Sybil attack, and Hello flood attack in WSN. And dynamic session keys between nodes are generated by using ID and bilinear principle, which reduces the interaction steps between nodes and energy costs. SVO logic reasoning and performance analysis show that the protocol can guarantee the security of WSN, and it meet the special application requirements of WSN.
  • LONG Xiao, WANG Jing-Cheng, DIAO An-Lei, HU Ting, ZHANG Shen-Wei, HE Jun, XU Chang-Jian
    Computer Engineering. 2012, 38(9): 125-126,130. https://doi.org/10.3969/j.issn.1000-3428.2012.09.037
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper designs an identity authentication scheme for Wireless Sensor Network(WSN) based on SHA-1 function. By using the feature of multi-mapping, it is widely used in terminal node of WSN which has limited resources. The hash operation is done on the random number sent by the platform and the terminal node’s key respectively, the results are integrated to be sinusoidal function variable which has larger value range, the variable is operated by sinusoidal projection, and a set of simpler projection values are obtained. Experiments on MICAz demonstrate that the scheme not only can improve the security of SHA-1, avoid traditional attack, but also can achieve better performance such as less memory space and energy.
  • MA Dong-Lan, ZHANG Jian-Zhong
    Computer Engineering. 2012, 38(9): 127-130. https://doi.org/10.3969/j.issn.1000-3428.2012.09.038
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The certificateless restrictive partially blind signature scheme does not satisfy the property of restrictiveness, and an account-holder can not be revealed when double-spending happens. In order to solve this problem, this paper proposes an improved scheme. In the new scheme, the two partial signatures of the signer in the original scheme can be bounded together, thus it can avoid forging valid signatures for any form of messages which does not meet the restrictiveness in the signature stage. Theoretical analysis results show that the improved scheme meets restrictiveness, partial blindness and unforgeability.
  • WANG Yong-Bing
    Computer Engineering. 2012, 38(9): 131-133. https://doi.org/10.3969/j.issn.1000-3428.2012.09.039
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Flexibility and practicability of hybrid proxy multi-signature are restricted bythe property that all of the signers produce signature together. In order to solve the problem, this paper proposes a (t, n) Threshold Hybrid Proxy Multi-signature(THPM) scheme combing with the threshold signature. A THPM is collaboratively generated by some original signers and some proxy signers on behalf of their original signers in the scheme. Only when t or more than t signers agree, the threshold number group members can produce a valid THPM in the signing group. Theoretical analysis proves that THPM not only has unforgeability, nonrepudiablity, distinguishability and quality of hybrid proxy multi-signature and threshold signature, but also is more efficient than original scheme.
  • LIN Li, GUAN De-Jun, XU Jian, ZHONG Ru
    Computer Engineering. 2012, 38(9): 134-137. https://doi.org/10.3969/j.issn.1000-3428.2012.09.040
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    From the attacker’s point of view, this paper sums up two features of attack on system and designs an automatic detection system for authentication protocol loophole. By target lookup algorithm, available resource lookup algorithm and protocol vulnerabilities searching algorithm, it finds a variety of ways to attack the authentication protocol to discover loopholes in the protocol. Needham-Schroeder, Neuman-Stubblebine and Otway-Rees protocol are used to do tests, whose results show that the system can correctly detect the vulnerability of the protocol, and simulate the attack methods.
  • SHI Qing-Li
    Computer Engineering. 2012, 38(9): 138-140. https://doi.org/10.3969/j.issn.1000-3428.2012.09.041
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to find buffer overflow vulnerability point accurately and rapidly, this paper proposes a method that can find buffer overflow vulnerabilities in binary file through code instrumentation. It uses plentiful functions PIN providing to make program analysis tool and saves information needed during program execution. When detecting memory access violation exception, it distinguishes what class of memory corruption and obtains memory corruption point and seeks illegal memory writing instruction to locate vulnerability. Example analysis shows that the method does not need source program, and has higher efficiency, it can locate popular buffer overflow vulnerabilities successfully.
  • ZHANG Xiao-Ping, ZHONG Cheng
    Computer Engineering. 2012, 38(9): 141-143,161. https://doi.org/10.3969/j.issn.1000-3428.2012.09.042
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Most ID-based signcryption schemes with public verifiability have security problems because of key escrow. So this paper proposes a public verifiable ID-based signcryption scheme without trusted Private Key Generator(PKG). A secret value chosen by signer is added as the partial private key, so that dishonest PKG can not forge the signature of user or decrypt and restore plaintext without authorization. Security analysis shows that the proposed scheme is secure against existential forgery on adaptively chosen massage and identity attack under the random oracle model, and it has the properties of confidentiality and traceability.
  • WANG E-Fei, ZHOU Ben-Dong
    Computer Engineering. 2012, 38(9): 144-147,165. https://doi.org/10.3969/j.issn.1000-3428.2012.09.043
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Most Identity-based Authenticated Key Agreement(ID-AKA) protocols have poor performance on security and the efficiency. To solve this problem, this paper presents an improved ID-AKA protocol. Security proof with the modular approach in the mBR model is provided, which means the new protocol satisfies perfect forward security, PKG forward security, resistant to key-compromise impersonation. Furthermore, the new protocol can satisfy resistance to leakage of ephemeral secrets. Compared with previous ID-AKA protocols which can satisfy the same security properties, the new protocol has higher efficiency.
  • WANG Dun-Wen, CA Juan, LI Hong-Yuan, LIU Guang-Jie
    Computer Engineering. 2012, 38(9): 148-150,169. https://doi.org/10.3969/j.issn.1000-3428.2012.09.044
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to use the digital fingerprinting to trace the user’s super distribution, this paper proposes a secure multimedia distribution scheme which supports the interoperable encryption and fingerprinting. It divides the 8×8 Discrete Cosine Transform(DCT) coefficients into four parts, and every part is encrypted with different methods. The digital fingerprints are directly embedded into midband coefficients of AC, and can be extracted after the decryption correctly. Experimental result verifies that the scheme gains the properties of imperceptibility, collusion security and robustness of fingerprints, and encryption algorithm is secure.
  • WU Qian, ZHOU Qiang
    Computer Engineering. 2012, 38(9): 151-152,173. https://doi.org/10.3969/j.issn.1000-3428.2012.09.045
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to ensure the workflow system secure and reliable, this paper introduces three relation elements of Target Case(TC), User Management(UM) and Target(T) in the conventional access control model based on role, designs the dynamic authorization mechanism, and builds a sort of security access model in workflow based on dynamic control mechanism, which ensures the secure running of the model by means of basic constraint relation and dynamic constraint conditions. The model is integrated with workflow engine component to provide the security authorization service for the application of independent security field. Application results show that the model can better separate the dynamic responsibility and reciprocal responsibility, make the binding of dynamic responsibility, and provide technical support for security access of system workflow.
  • QIN Shen, LAO Cui-Jin
    Computer Engineering. 2012, 38(9): 153-154,176. https://doi.org/10.3969/j.issn.1000-3428.2012.09.046
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problem of the current polymorphic worm feature extraction method does not handle noise, and the worm features which are extracted can not detect polymorphic worms effectively, this paper proposes an improved feature extraction algorithm. This algorithm extracts the features of the worm from the suspicious flow pool which have k worm series in n series by the Gibbs algorithm, then uses the method of color coding to improve the efficiency of algorithms’ run in the process of identifying the worm series. Simulation results show that this approach can reduce the time and space overhead. Compared with the existing feature extraction algorithms, this algorithm can effectively extract the polymorphic worm when there is noise in the suspicious pool.
  • BANG Chang-Yong, CHU Ti-Fei, GU Chun-Xiang, MI Shun-Jiang
    Computer Engineering. 2012, 38(9): 155-157,179. https://doi.org/10.3969/j.issn.1000-3428.2012.09.047
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    There are errors in analyzing the completeness of a cryptographic algorithm by statistical testing method. By using the symbolic computation software Mathematica 7.0, the polynomial expressions(in terms of the plaintext bits and key bits) of the output of 1~5 round of the block cipher LBlock are obtained. The result shows that any bit of the output of the 5th round of LBlock depends on at most 45 plaintext bits and 49 key bits, which means that 5-round LBlock is not complete.
  • TUN Xue-Lian, SUN Bing-Yu, LI Wen-Bei, ZHANG Ji
    Computer Engineering. 2012, 38(9): 158-161. https://doi.org/10.3969/j.issn.1000-3428.2012.09.048
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The methods for forecasting relief food, which are based on the experts and the prediction results, are influenced by the experience of the experts. Considering the advantage of rough set in dealing with the uncertain knowledge, this paper proposes a novel method that combines rough set with Case-based Reasoning(CBR) for forecasting relief food demand. This method utilizes discernibility matrix and attributes significance to reduce case attributes and get attribute weights. Experimental results prove that the proposed method can reduce the subjective effects and improve the predication accuracy.
  • TIAN Meng-Chu, CHEN Zhi-Min, WEI Xiu-Meng, ZHOU Qing, WANG Zhen-Li
    Computer Engineering. 2012, 38(9): 162-165. https://doi.org/10.3969/j.issn.1000-3428.2012.09.049
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Particle filtering is not precise and has weak robustnest, and it is not able to meet the requirement of fault diagnosis of temperature control system in power plant. To solve these problems, a new particle filtering algorithm based on Hybrid algorithm is proposed. The algorithm looks for satisfactory solution space with artificial fish swarm algorithm, later the particles move to the high likelihood region with Particle Swarm Optimization(PSO) algorithm. It raises the accuracy. Simulation results show that this algorithm has the high precision, strong robustness and it is suitable for fault diagnosis of temperature control system.
  • LIANG Jin-Quan, ZHOU Zhi-Beng, LI Meng, MO Guo
    Computer Engineering. 2012, 38(9): 166-169. https://doi.org/10.3969/j.issn.1000-3428.2012.09.050
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To solve the problem of robot path planning under complicated environment, a genetic operation based on key chains is proposed. Both retroversion and taboo strategy are incorporated into heuristic neighboring search to guarantee the feasibility of path. The memory requirement and computing cost are reduced by extracting key chain from original feasible route. The optimization performance is improved by locally adaptive mutation and crossover operator acting on key chains. Experimental results show new algorithm can plan the safety trajectory of robot under complex obstacle environment and overmatch other algorithms, as well as running time can meet the demand of practical applications.
  • BAI Qiu-Ju, ZHOU Zhao-Shan, SHU Xi-Jun
    Computer Engineering. 2012, 38(9): 170-173. https://doi.org/10.3969/j.issn.1000-3428.2012.09.051
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper designs a thenar palmprint recognition system. In the diagnosis of allergic diseases, thenar palmprint can be divided into four grades based on its special candidates. This system is developed mainly depend on the above, and its main function is able to show the grade of the thenar palmprint quickly in the case of the palm picture is given, and in order to achieve the purpose of supporting clinical experts to diagnose. The classification of Gray Level Co-occurrence Matrix(GLCM) and SVM are used in this system to realize the systematic. Test results show that this system can realize the feature extraction and classification of thenar palmprint, and the precision and effect can make demand of medical aid diagnosis and research.
  • DIAO Chuan-Min, MA Xiao-Hu
    Computer Engineering. 2012, 38(9): 174-176. https://doi.org/10.3969/j.issn.1000-3428.2012.09.052
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to achieve better Electrocardiograph(ECG) characteristics from high-dimensional data and realize accurate automatic ECG classification, a novel method for ECG multi-classification is proposed. This method uses Nonnegative Matrix Factorization(NMF) for data dimension reduction and conducts multi-classification by Support Vector Machine(SVM). In implementing the conversion of high dimension to low dimension, NMF retains the original information and supplies better eigenvectors, so it improves the classification accuracy. By testing four kinds of ECG from the MIT-BIH arrhythmia database, the total accuracy is up to 99%.
  • YANG Jia-Wen, SUN Ge-Meng
    Computer Engineering. 2012, 38(9): 177-179. https://doi.org/10.3969/j.issn.1000-3428.2012.09.053
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Multidimensional associative memory neural networks can be used for image recalling. When the image is corrupted by high noise, the recalling image is not clear using the traditional method. In order to make the recalling image clearer, one library vector made up of four column vectors is used in the recalling image to take the place of the other traditional library vectors made up of two column vectors. That is, a new matrix is formed by vertically dividing the mage matrix into four small matrices of the same order and vertically concatenating the four matrices in order. A column vector of the new matrix is regarded as a library vector. Numerical examples show that the restored image is clearer and the recalling process spends less time when the former library vector is used.
  • LIU Jian-Wei, CHE Guang-Hui, LUO Xiong-Lin
    Computer Engineering. 2012, 38(9): 180-182. https://doi.org/10.3969/j.issn.1000-3428.2012.09.054
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Gaussian process has the ability to deal with complicated problems such as high dimension, small samples, and nonlinearity. A new algorithm called confidence-weighted learning is proposed by introducing Gaussian process into the online classification. It maintains a Gaussian distribution over weight vectors and makes adjustment over the weight vector with every sample instance. By this way, it correctly classifies examples with a specific probability. Experimental results show that the confidence-weighted classifiers are always more accurate than the perceptron classifier in the simulated datasets, as well as in the real datasets.
  • HUANG Dun, TIAN Sheng-Wei, YU Long, FENG Guan-Jun
    Computer Engineering. 2012, 38(9): 183-185. https://doi.org/10.3969/j.issn.1000-3428.2012.09.055
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes the mechanism of sentence sentiment identification based on sentiment tagged word level constituents acquired by an automatic classifier. Basic set of eight sentiment types is selected for the test. The first part of the sentence is shielded if the sentence contains adversative conjunction. Conditional Random Field(CRF) model is used to tag the words with the sentiment types automatically, and referring to the words of automatic tagging, giving a score to each sentiment type of a sentence. The highest score of sentiment type is chosen to candidate sentiment type. According to the number of negative components in sentences is oven or odd, and the final sentiment type of sentence is revised. Experimental results show that the method on sentence sentiment analysis achieves good effect.
  • CHEN Jian-Hua, WANG Chi-He, JIANG Yun
    Computer Engineering. 2012, 38(9): 186-188,192. https://doi.org/10.3969/j.issn.1000-3428.2012.09.056
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes a syntaxic feature selection algorithm based on category discrimination degree and correlation analysis. The algorithm uses discrimination degree to extract the features that reveal larger differences among categories to reduce the sparsity of feature spaces, and emploies correlation analysis of features to measure relativity between features and categories and redundancy among features, so it can acquire the feature subsets which are more representative and have no redundancy between each other. Experimental results show that the proposed algorithm can improve the performance of the classifier effectively.
  • ZHANG Jian-Meng, WEI Lin-Feng, LIU Zhi-Jiang, HONG Peng
    Computer Engineering. 2012, 38(9): 189-192. https://doi.org/10.3969/j.issn.1000-3428.2012.09.057
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to improve accuracy and reliability of the detection based on a single feature, a detection algorithm based on Bayesian network fusing multiple features parameters is proposed. The algorithm locates driver’s eyes and mouth, creates head revolution pattern, which is based on the triangle composing of two eyes and mouth, then inputs all the extracted character parameters to Bayesian network and makes a logic judgment, a timely warning is given when detecting drivers in an abnormal driving status. Experimental results show that the algorithm for detecting drowsy or losing attention has high accuracy.
  • HUANG Cun-Dong, LIU Ren-Jin, YANG Sai-Chun
    Computer Engineering. 2012, 38(9): 193-196. https://doi.org/10.3969/j.issn.1000-3428.2012.09.058
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    A new approach of face recognition in videos, named FFME, is proposed. Faces are divided into different regions. Different local feature descriptors, which are selected against the variation of some region, are combined for face representation. K-NN model is built for feature vectors classification. A sum rule is followed by FFME to combine the individual classification results. Manifold is introduced for constructing a set of reference face images to re-rank the top candidate images, which is approved later to enhance the face recognition rates. Experimental results on the video-based face database Mobo and Honda/UCSD show that FFME is superior to other methods such as principal components analysis, linear discriminant analysis, hidden Markov model, locally linear embedding and manifold-manifold distance.
  • BO Feng, WANG Jian-Dong, GU Ji-Wei, NIU Ben
    Computer Engineering. 2012, 38(9): 197-198,201. https://doi.org/10.3969/j.issn.1000-3428.2012.09.059
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The high dimensionality of the data samples often makes the data mining or pattern recognition tasks intractable, through analyzing both the within-class distance and between-class distance, it presents a fast feature ranking framework, from which the computationally expensive feature decomposition is avoided. Two similarity measures of within-class and between-class similarity and K nearest neighbor similarity are employed to derive efficient feature selection algorithms. Experimental results demonstrate that these algorithms have higher classification precision.
  • XU Dun-Gong, DING Chun-Feng, SU Hai-Bin, WANG Ting-Ling
    Computer Engineering. 2012, 38(9): 199-201. https://doi.org/10.3969/j.issn.1000-3428.2012.09.060
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the shortage of Gradient Vector Flow(GVF), a new moving object extraction algorithm based on improved GVF is proposed. A rectangle limiting the range of target is established by experience and displacement equation. Initial contour according to rectangle range is obtained by rules. The algorithm of GVF is improved, which is used to calculate fine contour. Experimental results in sequence display the algorithm is simple and rapid.
  • SHANG Li-Beng, LIU Jian
    Computer Engineering. 2012, 38(9): 202-204,207. https://doi.org/10.3969/j.issn.1000-3428.2012.09.061
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to improve the accuracy of traditional linear dynamics method, an automated Empirical Mode Decomposition(EMD) signals recognition method is proposed. The method which combines the nonlinear dynamics with the time-frequency analysis is used to study signals. In order to effectively restrain noise and decompose the signals, EMD is employed to preprocess the sample data, and the approximate entropy of some important intrinsic mode functions can be obtained. Support Vector Machine(SVM) is adopted to achieve the optimal classification. Experimental results show the correct classification rate can be more than 90%, which proves the reliability and veracity of the method.
  • DIAO Lv, SU Yi-Dan, QIN Hua
    Computer Engineering. 2012, 38(9): 205-207. https://doi.org/10.3969/j.issn.1000-3428.2012.09.062
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To further improve the efficiency of training on large sets in Relevance Vector Machine(RVM), this paper proposes an optimized algorithm based on rapid estimation. That is preprocessing the training set and excluding the non-relevance vector with iterative estimation integrated by threshold factor and maximum reduction toplimit, then decreases the sample scale and reduces the training time. Experimental results on UCI datasets illustrate that this algorithm has faster training rate while maintaining training precision.
  • SHEN Li-Jun, LIU Li, LIU Dui, CHEN Yu-Ting, TIAN Beng-Beng
    Computer Engineering. 2012, 38(9): 208-210. https://doi.org/10.3969/j.issn.1000-3428.2012.09.063
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The task scheduling method built-in cloud computing environment is inefficient. A method based on improved immune evolutionary algorithm is proposed for task scheduling, which roots from artificial immune evolutionary theory to solve global optimize task scheduling on cloud computing. The improved immune evolutionary algorithm(Particle Immune Evolutionary Algorithm)PIEA introduces Particle Swarm Optimi- zation(PSO) into immune evolutionary algorithm. PIEA improves the optimization ability compared with traditional immune evolutionary algorithm, and avoids local Optimization, the convergence of this method is better, and time consuming of task scheduling is reduced. The CloudSim simulation platform is chosen, and results indicate that PIEA can provide efficient task scheduling strategy.
  • MI Xiao-Chi, GU Shi-Jie, LIU Wen-Yi
    Computer Engineering. 2012, 38(9): 211-213. https://doi.org/10.3969/j.issn.1000-3428.2012.09.064
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In this paper, a new localized energy active contour model combined with multiple features is proposed for image segmentation, especially for those images with the inhomogenous appearance features in foreground and background. It reformulates the localized segmentation energy to enhance its robustness to initialization. By introducing more fine 8-d HSV color space and structure tensor texture features of transform domain, it is able to segment images with weak object boundaries, images with visually similar color properties in foreground and background. Experimental result shows that the proposed model is exemplified on various images objects and its superiority over state of the art variational segmentation techniques is demonstrated.
  • CHEN Jun, HU Fu-Jiao
    Computer Engineering. 2012, 38(9): 214-216,233. https://doi.org/10.3969/j.issn.1000-3428.2012.09.065
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To overcome the color distortion and computational time-consuming problems with traditional RGB color space model, this paper proposes a method based on YUV color space model for creating High Dynamic Range(HDR) color images. This method uses a series of digital images with different exposure to generate a HDR color image. Experimental results show that the method is more efficient in computational speed and has better color image quality in comparison.
  • ZHANG Li-Yi, CHEN Gen-Fang, DU Xuan, HU Hai-Ou, DIAO Qiong-Di
    Computer Engineering. 2012, 38(9): 217-219,236. https://doi.org/10.3969/j.issn.1000-3428.2012.09.066
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to acquire lyrics of numbered-notation score, this paper proposes an scheme of image segmentation and lyrics extraction for Chinese numbered-notation score. During the image segmentation stage, it separates lyrics from music and title regions utilizing the projection and high-level domain knowledge of numbered-notation score. During the lyrics extraction stage, it designs Bounding Box(BB) algorithm to extract lyrics, merges the boxes belonging to one character, distinguishes punctuations from lyrics and stores them separately. Experimental results show that this scheme can extract lyrics effectively, and the average rate of lyrics extraction reaches 97.48%.
  • ZHANG Ya-Juan, LI Ze-An, CHENG Chen
    Computer Engineering. 2012, 38(9): 220-222,239. https://doi.org/10.3969/j.issn.1000-3428.2012.09.067
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Tranditional triangular mesh models exist noise and disturbance problems. Aiming at this problem, a modified algorithm is presented in this paper. It chooses different filtering factor in the characteristics area or flat area of mesh model by using the method of setting threshold value. At the same time, with the nature of Gaussian functions, structuring trend parameters according to the local shape information of each vertex. In addition, it takes the change rate of the vertex normal as the adjustment coefficient that can adjust the vertexes. Experimental result shows that the algorithm has less number of iterations, smaller contraction of surface area, higher efficiency of implementation and self-adaptive characteristics.
  • A Li-Mu-?Sai-Mai-Chi, DU Pei-Jun, LIU Sai-Cong
    Computer Engineering. 2012, 38(9): 223-225,243. https://doi.org/10.3969/j.issn.1000-3428.2012.09.068
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problem of large computing in maximum 2D entropy based image segmentation method, this paper proposes a maximum 2D entropy image segmentation algorithm based on artificial bee colony optimization. Artificial bee colony algorithm has certain advantage in convergence speed, prevents local optimization, and has few control parameters. Using these advantages, the best 2D threshold of maximum 2D entropy method is considered as nectar, and artificial bee colony optimized maximum 2D entropy method is used to segment images. Experimental result shows that, compared with other methods, constriction of this method is quicker, stability is better and resistance to the noise is stronger.
  • YIN Xiao-Hui, ZHANG Bao-Ju, WANG Wei, LEI Qing
    Computer Engineering. 2012, 38(9): 226-227,247. https://doi.org/10.3969/j.issn.1000-3428.2012.09.069
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to improve the quality of the compressed image, an compressed sensing method based on improved layered Discrete Cosine Transform(DCT) is proposed. It only measures the high-pass coefficients of the other layers, and preserves the top layer coefficients. For the reconstruction, high-pass coefficients can be recovered by the measurements. The image can be reconstructed by the inverse DCT transformation. Simulation results demonstrate that, at the same compression ratio, the peak signal to noise ratio of this method is higher than method based on layed DCT.
  • ZHENG Jing-Bi, WANG Chun-Xia
    Computer Engineering. 2012, 38(9): 228-230,251. https://doi.org/10.3969/j.issn.1000-3428.2012.09.070
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Considering the high computational complexity, low robustness, and inefficiency in solving image matching and stitching seam by traditional method for generating global view face texture image, an improved method is proposed in this paper. The face images to be spliced are normalized. An exact match of images is achieved according to gray scale image matching algorithm based on the column and feature vectors. The double fusion algorithm is used to smooth the overlapping area. Experimental results show that this method can eliminate false matches validly and maintain good image stitching accuracy with a high speed.
  • XU Cong-An, JIAN Chao, SUN Wei-Chao, GU Xin-Feng-
    Computer Engineering. 2012, 38(9): 231-233. https://doi.org/10.3969/j.issn.1000-3428.2012.09.071
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    For solving the robust disadvantage of the existing Constant False Alarm Rate(CFAR) detectors, a new CFAR detector based on Anderson-Darling(AD) test called AD-CA is proposed. It takes the critical value of AD test through Monte Carlo simulation, then deletes the unwanted samples. It is evaluated through simulation in various environments. Simulation result shows that, in homogeneous background, the performance of AD-CA is as good as OS; In multiple targets background, compared with the OS, the detection performance of AD-CA improves obviously. Especially when the value of interfering targets exceeds N–k, it also keeps good performance.
  • LONG Wen, QIN Gao-Yu
    Computer Engineering. 2012, 38(9): 234-236. https://doi.org/10.3969/j.issn.1000-3428.2012.09.072
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Under the conventional control, the temperature is not easy to control in the process of beer yeast spread cultivation. Aiming at this problem, a Proportional-integral-differential(PID) control method based on hybrid Genetic Algorithm(GA) is proposed. A gradient descend operator is introduced into the GA. Gradient descend method is carried on search by certain probability for certain elitists of every generation to strengthen the ability of local search. Experimental results show that the method can ensures fast response performance of system and it has not steady-state error.
  • LIU Ji-Yao, LIU Lei-Bei, YIN Shou-Yi, WEI Shao-Jun
    Computer Engineering. 2012, 38(9): 237-239. https://doi.org/10.3969/j.issn.1000-3428.2012.09.073
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In this paper, a heterogeneous multi-core System on a Chip(SoC) based on multi-layer AHB architecture is proposed. The processor subsystem consists of an ARM processor and two DSP processors. Therefore both control-intensive tasks and computing-intensive tasks are executed efficiently. Both distributed memory and shared memory system are employed in order to achieve the trade-off between data integration and program parallelism. The communication mechanism is based on switching matrix architecture. As long as there is no competition for one certain slave, different masters are able to access the bus simultaneously. Experimental result shows that this system has low resource consumption and latency. It is able to achieve high bandwidth.
  • TUN Chang-Rui, KONG Chao, JI Yang-Hui, CEN Fan, CA Hui-Zhi
    Computer Engineering. 2012, 38(9): 240-243. https://doi.org/10.3969/j.issn.1000-3428.2012.09.074
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Wideband phased array radar requires that the corresponding signal processing system can instantaneously process wideband data under stringent conditions. Aiming at this problem, this paper presents a high performance signal processing system based on VPX serial bus. The system achieves flexible transferring capability through the combination of high speed serial communication protocols and excellent processing capability through Field Programmable Gate Array(FPGA) technology and PowerPC technology. Application result shows that the system supports wider signal band and higher power and can meet the demand of wideband phased array radar.
  • XUE Chao-Fan, ZHANG Cheng-Bing
    Computer Engineering. 2012, 38(9): 244-247. https://doi.org/10.3969/j.issn.1000-3428.2012.09.075
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to extend the application field of chip and improve the DSP ability of chip, this paper proposes a design of supporting zero-overhead loop for the Micro Controller Unit(MCU). Based on the character of frequent loops in the Digital Signal Processing(DSP) program, a special hardware module for handling the loops is designed to eliminate the stall of pipeline caused by the loop. By analyzing the original structure of MCU, especially instruction fetch unit, the different treatments are taken from the other branch instructions for the loop instructions. The structure of the MCU is modified as little as possible to support zero-overhead loop. Performance analysis results show that the MCU changed reduces the loop’s execution cycles.
  • TAO Wu-Jian, SHI Xiao-An, ZHOU Xin-She-
    Computer Engineering. 2012, 38(9): 248-251. https://doi.org/10.3969/j.issn.1000-3428.2012.09.076
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Existing resource management methodologies for embedded real-time applications have limitations in handling dynamic environments, the system cannot ensure the Quality of Service(QoS). According to requirements of new generation typical dynamic real-time system, the eCos operating system of Red Hat corporation is improved by applying control theory-based adaptive real-time resource management. An Embedded Adaptive Real-Time Resource Management(EARTRM) is designed and developed, adding a control theory-based real-time scheduling framework, a RM Scheduling, a real-time task API, and a system monitor. Experimental result shows that this embedded adaptive real-time resource management has more precise and efficient QoS and guarantees the real-time dynamical requirements of resource and the robust performance of real-time tasks.
  • DING Yuan-Meng, LI Hua-Fang, LIU Chun-Ling, ZHOU Hua-Yu
    Computer Engineering. 2012, 38(9): 252-254. https://doi.org/10.3969/j.issn.1000-3428.2012.09.077
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper designs a transmitting system which is available in satellite communication. This system is based on Virtex-5 Field Programmable Gata Array(FPGA). It consists of PCI Express bus and optical communication module based on Aurora protocol. This system has the character of sending and receiving simultaneously for bidirectional I/O. Test result shows that the system achieves that the eight channels of PCI Express, and optical communication can works in real time data communication at a high-speed of 1.25 Gb/s. The system can effectively simulate the satellite data download as well as the remote control of the ground stations towards satellite.
  • WANG Tian-Bao, LEI Gao, ZHONG Er-Shun
    Computer Engineering. 2012, 38(9): 255-257,261. https://doi.org/10.3969/j.issn.1000-3428.2012.09.078
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In terms of various Geographic Information System(GIS) Web services generated by different sources and implemented by diverse technology, this paper proposes a new Web Geographic Information System(WebGIS) development framework under perspective of client terminal. It makes the sorting method for the Geospatial Web Service(GWS) from the angle of consuming service, and designs the corresponding model and architecture. The development framework is studied and applied in the actual cases, which shows that the new sorting method and framework are effective and practical.
  • SUN Wen-Qing, HU Ling-Min
    Computer Engineering. 2012, 38(9): 258-261. https://doi.org/10.3969/j.issn.1000-3428.2012.09.079
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The existing anti-collision algorithms in Radio Frequency Identification(RFID) system application have the power consumption and high cost. In order to solve this problem, this paper proposes an anti-collision algorithm for the multi-reader based on schedule mode. The improved algorithm is proposed based on the anti-collision of scheduling, which adoptes the conception of cluster headers, the method of reservation and dormancy count, to realize the zero collision in the phase of transmission. Simulation results show that the proposed algorithm gains the advantage over PULSE protocol, which avoids the signal interference significantly, improves the reading efficiency, and meanwhile reduces the power consumption. With the number of readers increased, the superiority of the performance is more obvious.
  • BANG Hui, HE Yi-Hui, JIANG Feng, WANG Yong
    Computer Engineering. 2012, 38(9): 262-264,267. https://doi.org/10.3969/j.issn.1000-3428.2012.09.080
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In this paper, the tactical path planning method for Computer Generated Force(CGF) Agent in urban combat simulation is proposed. It analyzes the tactical requirement of Agent path planning, and the mathematics model of path optimization based on the distance cost and tactical cost is established. A Probabilistic Roadmap Method(PRM)-based tactical path planning framework is presented. Experimental results show that the average planning time is 72.94 ms, the maximum and minimum planning time are 110 ms and 15 ms.
  • HUI Jin-Fen, SONG Bao-Wei, MAO Zhao-Yong
    Computer Engineering. 2012, 38(9): 265-267. https://doi.org/10.3969/j.issn.1000-3428.2012.09.081
    Abstract ( )   Knowledge map   Save
    In order to realize the fusion of information of multiple sources, the model of Bayes reliability analysis based on the multistage experiments is established, which classifies the test data according to the test stages, and making the previous stage of the test data as a priori information for the later stage. Followed by the recursive, the multi-stage fusion can be achieved based on the Bayesian method. Simulation results show that the model is simple, and can effectively fuse the multiple sources information, which realizes Bayesian reliability evaluation based on the small sample.
  • MAO Hua, DIAO Xiao-Na, MAO Xiao-Liang
    Computer Engineering. 2012, 38(9): 268-270,274. https://doi.org/10.3969/j.issn.1000-3428.2012.09.082
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In hazardous materials transportation, to make the dangerous up to the lowest, this paper proposes a minimal risk and maximal flow algorithm. It presents an idea that all of the affected factors are considered as one characteristic risk ratio. Using the method of risk ratio, it obtains a path of minimal risk ratio. Along with this path, the feasible flows are increased. This realizes an algorithm of minimal risk in hazardous materials transportation. Numerical example shows that compared with the classical algorithms relative to minimal cost flow and minimal cost path, this new algorithm can make the complexity lower and take much smaller spaces.
  • DIAO Yuan, CENG Guang-Yu, WANG Wei, CUI Chen, GAO Hong-Bo
    Computer Engineering. 2012, 38(9): 271-274. https://doi.org/10.3969/j.issn.1000-3428.2012.09.083
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    For traditional Virtual Instruction Set Architecture(V-ISA) can not satisfy both brevity and high-efficiency well, this paper proposes a Virtual Instruction Set Architecture(V-ISA) design method for firmware code analysis. A muti-target firmware code analysis platform is designed, and on the basis of virtual hardware configuration that can be configured, a method for developing virtual instruction set is proposed that it first build the minimal complete instruction set and then expands it. Experimental result shows that this method can reduce translated-code expansion rate and get the simulation costs have a reduction of 19%~35% compared with the conventional one.
  • CHEN Jia, KUANG Zhi-Feng, LI Min
    Computer Engineering. 2012, 38(9): 275-277,281. https://doi.org/10.3969/j.issn.1000-3428.2012.09.084
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Twitter as a research subject, this paper proposes a Social Network Analysis(SNA) method. Respectively, considering concerns relations between users and information transmission perspective from the micro-blog, it uses SNA to quantitatively analyze social network structure, studies of its social network density, centrality, position and role. Analysis results show the method can avoid difficult problems of information fragmentation, negative news explosive spreading, and be applied to monitor public opinion on national security.
  • BO Chao, ZUO Kai-Zhong, GUO Liang-Min, LUO Yong-Long
    Computer Engineering. 2012, 38(9): 278-281. https://doi.org/10.3969/j.issn.1000-3428.2012.09.085
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to overcome the high complexity of tag and resynchronization between tag and database, a new mutual authentication protocol is proposed based on Exclusive OR(XOR) operation. The new protocol can deal with the synchronization of data update by changing the update order. As only XOR and random number generator are required to be computed by tags, it is very suitable for low-cost Radio Frequency Identification(RFID) system compared with hash-based authentication protocols. Experimental results show that this protocol can guarantee the safety of data transmission, and the hardware cost and computation spending is low.
  • TU Yong, LONG Jian-Min, CHAN Zheng, LIU Xiao-Nan
    Computer Engineering. 2012, 38(9): 282-284,287. https://doi.org/10.3969/j.issn.1000-3428.2012.09.086
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problem that the number of threads is not matched when migrating Compute Unified Device Architecture(CUDA) programs to other heterogeneous many-core architecture, the paper proposes a hierarchical thread mapping model. In the first level, the model maps the host threads and device threads of CUDA to the target platform’s master core and slave core array respectively. In the second level, the model removes synchronization operations of Cooperative Threads Array(CTA) threads by thread loop, and maps the whole CTA to a slave core. Experimental results show that the model allows CUDA programs to be implemented effectively in other heterogeneous many-core systems.
  • JIANG Xiao-Lu, LIU Yuan
    Computer Engineering. 2012, 38(9): 285-287. https://doi.org/10.3969/j.issn.1000-3428.2012.09.087
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to improve the efficiency of collision detection in complex scence, this paper makes some optimizations of the traditional collision detection algorithm based on hybrid bounding box. It improves the data structure of hierarchical bounding box and adopts the concept of temporal-spatial coherence. The box tree falls into upper and lower levels. The upper layer uses sphere, and the lower layer uses Axis-Aligned Bounding Box(AABB). It builds hybrid bounding box to achieve fast collision detection. It uses temporal-spatial coherence to simplify the search processes. This speeds up the collision detection. Experimental results show that this algorithm is more effective than traditional collision detection algorithm based on hybrid hierarchical bounding box.
  • QIU Yun-Fei, CHENG Liang
    Computer Engineering. 2012, 38(9): 288-290. https://doi.org/10.3969/j.issn.1000-3428.2012.09.088
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Against the problem that Topic Detection and Tracking(TDT) can not deal with the short message texts on microblog which have strong randomness and non-standard terms effectively, a detecting method of sudden topics on microblog based on the dynamic sliding window is proposed. It includes the use of windows to extract the information with potential sudden, the use of normalized Term Frequeney-Inverse Doeument Frequeney(TF-IDF) function which is combined with semantic to compute feature weight and build Vector Space Model(VSM), the use of the main idea of Single-Pass clustering algorithm and then improving it in order to generate the final clustering. Experimental results show that the algorithm has an accurate result in sudden topic detection.
  • XU Zhi-Gong, LIU Jin-Jun, DIAO Sheng-Hui
    Computer Engineering. 2012, 38(9): 291-292,封三. https://doi.org/10.3969/j.issn.1000-3428.2012.09.089
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    With redundancy detection, this paper proposes a strategy of disk live migration. The strategy maintains data consistency of virtual machine disk on source and destination using cycle synchronization. In the non-synchronous period, the strategy marks the position of dirty blocks by bitmap. In the synchronous period, the strategy detects Hash values of dirty blocks between source and destination, and synchronizes inconsistent blocks to the destination node. After data consistency, the strategy triggers live migration. Experimental results show that the strategy can reduce the amount of disk migration, and decrease the demand for high bandwidth, the migration time is closed to the shared storage approach.